By David N. Leff

Treating Parkinson¿s disease with an antibiotic may seem like a prescription-filling mistake. In fact, the antibacterial drug in question, minocycline, has been tried in recent years on Huntington¿s disease and Alzheimer¿s as well, with interesting results.

To be sure, in all three cases ¿ PD, HD and AD ¿ scientists administered the minocycline not to human patients, but to mouse models of these neurodegenerative diseases. The latest such counterintuitive experiment is reported in today¿s Proceedings of the National Academy of Sciences (PNAS), dated Dec. 4, 2001. Its title: ¿Minocycline prevents nigrostriatal dopaminergic neurodegeneration in the MPTP model of Parkinson¿s disease.¿ Its senior author is neuroscientist Steven Paul, group vice president of Lilly Research Laboratories, of Indianapolis.

People, not mice, were the first mammals to display PD signs and symptoms in response to MPTP. That acronym stands for 1-methyl-4-phenyl-1,2,3,6,-tetrahydropyridine, and it surfaced over 20 years ago as a PD-causing toxin. That was when a bunch of heroin addicts in California tried shooting up a new, synthetic designer heroin, and ended up ¿ oops! ¿ getting MPTP instead. It sent four of them to the hospital with irreversible manifestations of Parkinsonism.

The toxin has since become a favorite resource for creating mice that faithfully simulate the chronic clinical hallmarks of PD. MPTP selectively destroys dopamine neurons in the substantia nigra. This brings about a Parkinson¿s-like syndrome in mice, monkeys and man.

In their PNAS paper, Paul and his co-authors report that minocycline, a semisynthetic form of tetracycline, protects brain cells from destruction by PD-modeling MPTP, namely loss of dopamine neurons in the substantia nigra, decreased striatal dopamine levels, and consequent extrapyramidal motor dysfunction.

Dopamine is the key neurotransmitter progressively lost in Parkinson¿s disease. Hence, the principal therapeutic for the disorder is levodopa (L-dopa), an anti-Parkinsonian agent that is converted to dopamine in the brain. L-dopa initially boosts dopamine function and temporarily mitigates the trembling, shaking, rigidity and decreased movement spontaneity of PD. But the drug fails to halt disease progression, loses efficacy and inflicts side effects, such as difficulty performing voluntary movements.

High-Dose Minocycline Worked ¿ But No Cigar

Paul and his co-authors created their MPTP mouse model of PD, ¿using what is described in the literature. It¿s where you inject MPTP, which is neurotoxic,¿ he told BioWorld Today. ¿In our mice, we did this in a fairly acute fashion. We administered the MPTP by injection four times a day on a single day. Then, we pretreated the animals a bit with oral minocycline ¿ 120 milligrams per kilogram ¿ two days before the initial MPTP administration. A week or two weeks later, we looked to see if we got dopamine neuron degeneration. We didn¿t measure behavioral effects. Mice are not as easy to look at the behavioral affects. Rats are easier.

¿We measured dopamine levels in the striatum. And of course with MPTP lesioning, they were quite reduced, and we found an attenuation of that with the higher dose, particularly of minocycline. We looked at catecholamine levels in the striatum, dopamine and its metabolites. And we counted tyrosine-hydroxylase-positive nerve cells. They¿re dopamine neurons in the substantia nigra that project to the striatum. Those are the neurons that degenerate in PD. We found they were not as dramatically reduced in the animals that we pretreated with the minocycline antibiotic.

¿Some Finnish authors,¿ Paul recalled, ¿had previously reported work showing minocycline to be effective in two models of stroke ¿ focal and global ischemic brain injury. And some work had been done in an animal model of Huntington¿s disease. We noted a whole body of literature showing minocycline as having anti-inflammatory effects, quite apart from its antimicrobial action.

¿Our PNAS paper,¿ he continued, ¿demonstrates that minocycline has neuroprotective effects, now extended to our MPTP model of PD.

¿We think its significance is that mechanistically the paper suggests a novel drug that¿s effective in this model. Now we would like to know what the exact mechanism is. What is minocycline doing? What processes is it affecting? So we¿re trying to pin down exactly how minocycline may be working. And we think if we find that out we may be able to craft a new molecule, a new drug, that might be very effective in preventing PD.¿

Meanwhile, he added a strong caveat. ¿We want to be very cautious, because we are not in any way, shape or form advocating the use of minocycline to treat PD. In fact, in the animal model, the effects of that antibiotic in protecting neurons in vivo were only observed at very high doses of minocycline. And we believe that¿s because minocycline itself, as we showed, doesn¿t get into the brain very well, whereas MPTP does.

¿Now there¿s another thing about tetracycline you have to be careful about,¿ Paul added. ¿You wouldn¿t necessarily want to use an antibiotic to treat Parkinson¿s disease or, for that matter, any other neurodegeneration because of drug resistance.¿

From PD To Hyperbilirubinemia Outlook

Since submitting his paper to PNAS last August, Paul related, ¿We¿re trying to work out the mechanism of how minocycline might be protecting neurons, looking biochemically to see if we could figure out how it may be working here.

¿We¿ve also looked at some other models of neurotoxicity,¿ he continued, ¿specifically a bilirubin model. It turns out that if we put bilirubin, which is a breakdown product of hemoglobin, on neurons, it will also kill them. And we find that minocycline blocks bilirubin as well. It¿s unrelated to PD, but it¿s another model of neurodegeneration.

¿In vivo, it works also in an animal model of hyperbilirubinemia brain damage ¿ the Gunn rat model. It has a naturally occurring genetic mutation in one of the enzymes that conjugates bilirubin. If those animals are missing that enzyme completely ¿ as in some of the homozygous rats ¿ what we get is a very high level of circulating bilirubin. And that causes the cerebellum to become very hypoplastic, to shrink. We got pretty good protection when we treated the little rat pups with minocycline. So we¿ve looked at some other models of neurodegeneration, which is where my group and I are going.¿